This study was performed within the European Network for the Investigation of Gender Incongruence (ENIGI) study, a multicentric prospective study resulting from the collaboration of four gender clinics (Ghent, Amsterdam, Florence and Oslo), aiming to evaluate the effect of GAHT on several issues, including dermatological changes [15, 16]. Study design and methods have been extensively described elsewhere [15, 16]. In particular, for the present sub-study, data from Amsterdam, Ghent and Florence were selected. Persons were eligible for inclusion in the study if they were at least 18 years old, had a diagnosis of gender dysphoria based on formal classification criteria [17], and were about to start GAHT. Exclusion criteria included previous or current use of any hormonal preparation, illiteracy/mental retardation, absence of at least a 12-month follow-up, and treatments or disorders affecting dermatological aspects (thyroid disorders, Cushing syndrome, cirrhosis, chronic kidney failure, previous use of antiandrogens or anabolic steroids). Written informed consent was obtained from participants according to institutional guidelines.

The final study sample consisted of 484 participants (291 trans AFAB people and 193 trans AMAB people), enrolled before starting GAHT and evaluated prospectively at baseline (T0), 6 months (T1) and 12 months (T2) after GAHT prescription. All participants reported a binary gender identification and requested a standardized GAHT [2, 4].

All trans AFAB people received T treatment to obtain virilization [1]. Since full virilization was requested, T was administered at the same dosages used in hormone replacement treatment of hypogonadal cisgender men (aiming to achieve T values in the normal male range) [1]. More specifically, intramuscular (i.m.) injections of T undecanoate 1000 mg (the second injection repeated after six weeks, then after 12 weeks) or combination of T esters (250 mg/14 days) was prescribed to 142 and 73 participants, respectively, while transdermal T gel (50–60 mg/day) was prescribed to 76 participants.

As requested, treatment of trans AMAB people consisted of both estrogens and antiandrogens, in order to obtain a full feminization and de-masculinization [1]. All patients received cyproterone acetate (50 mg/daily) combined with estradiol valerate (2 to 6 mg/daily) in 97 participants, estradiol patches (50 to 100 mcg/24 h twice weekly) in 85 participants and estradiol hemihydrate gel (2–3 mg/daily) in 11 participants. Despite different hormonal preparations, in trans AMAB people, target estradiol levels were maintained between 100 and 200 pg/ml with T levels < 50 ng/dL [1].

All participants completed a self-reported questionnaire with questions regarding medical history, smoking status, previous and current hormonal treatment, medications and dermatological history.

At baseline and during each visit, patients underwent a physical examination including measurement of height, weight, body mass index (BMI) and dermatological assessment. The degree of hair growth was assessed by expert endocrinologists using the modified Ferriman and Gallwey (FG) scoring system [18,19,20]. This score is largely validated only in cisgender population and considers the presence and distribution of terminal hair in nine sites (lip, chin, chest, upper back, sacroiliac region, upper abdomen, lower abdomen, arms and medial thigh). Each site is rated from 0 (no growth of terminal hair) to 4 (very dense hair growth), with a total score ranging from 0 to a maximum of 36. Due to frequent hair removal, in trans AMAB people, the score was adjusted by asking participants to describe medium hair density during the last 2 weeks before depilation or waxing.

Occurrence and pattern of AGA was assessed with the Norwood Hamilton (NH) score [21]. This scale contains seven stages describing typical sequences in the development of male pattern baldness, with higher scores indicating more severe hair loss (for details of different grades, see [22]).

A subsample of patients from Ghent and Florence (AFAB n = 71, AMAB n = 26) underwent acne assessment. Acne evaluation was performed on the face and back using the Global Acne Grading Scale (GAGS) [23]. GAGS evaluates six locations of the face and chest/upper back, rating each region with a score depending on the type of lesion (no lesion = 0, one comedone = 1, one papule = 2, one pustule = 3, one nodule = 4). These grading scores are then multiplied by a specific factor for each location (forehead ×2, right cheek ×2, left cheek ×2, nose ×1, chin ×1, chest and upper back ×3). After that, the sum of local scores provides the global score (ranging from 0 to 52). Acne severity is graded as mild (global score = 1–18), moderate (global score = 19–30), severe (global score = 31–38) and very severe (global score > 39).

Venous blood samples were obtained at baseline, 6 and 12 months after overnight fasting, independently from the time of hormonal treatment administration. Samples were analyzed at the local laboratory. Biochemical determinations included T, estradiol, luteinizing hormone (LH) and follicle-stimulating hormone (FSH) levels.

In Amsterdam, serum estradiol levels were measured through a competitive immunoassay (Delfia; PerkinElmer, Wallac Oy, Turku, Finland) with an interassay coefficient of variation (CV) of 13% and a lower limit of quantification (LOQ) of 20 pmol/l until July 2014. Thereafter, liquid chromatography tandem mass spectrometry (LC–MS/MS) was used, with an interassay CV of 7% and an LOQ of 20 pmol/l. Until January 2013, serum T was measured using a radioimmunoassay (RIA) (Coat-A-Count, Siemens, Los Angeles, CA; interassay CV 7–20%; LOQ 1 nmol/l). After January 2013, T levels were determined using a competitive immunoassay (Architect; Abbott, Abbott Park, IL) with an interassay CV range of 6% to 16% and an LOQ of 0.1 nmol/l.

In Ghent, until March 2015 estradiol was measured using E170 Modular (E2 Gen II, Roche Diagnostics, Mannheim, Germany). After that, E170 Modular (E2 Gen III, Roche Diagnostics) with an interassay CV of 3.2% and an LOQ of 92 pmol/L was used. T levels were measured using E170 Modular (E2 Gen II) with an interassay CV of 2.6% and an LOQ of 0.4 nmol/l.

In Florence serum estradiol was measured using electrochemiluminescence immunoassay (E801, Roche Diagnostic) with an interassay CV of 3.6% and an LOQ of 91.8 pmol/l. T levels were measured through E801 (Roche Diagnostics) with an interassay CV of 6.8% and an LOQ of 0.416 nmol/l.

Data were analyzed using IBM SPSS 26.0 (SPSS, Chicago, IL, USA). Sample characteristics are presented as percentages with categorical variables, or as mean ± SD for continuous variables. Data were checked for normal distribution by means of the Kolmogorov Smirnov Statistic. For the assessment of between-group differences (trans AFAB vs. trans AMAB people), a χ2 and an independent measure t test were applied for categorical and continuous variables, respectively. Differences between groups were evaluated in a multivariate model (adjusting for the relevant clinical confounders) by means of an analysis of covariance (ANCOVA) with post hoc Bonferroni test. Statistical significance was determined at p < 0.05.

To evaluate changes in dermatological outcomes across time, a mixed linear model was applied to the outcome variable, with visit (number of months of GAHT) as the fixed factor and with a random intercept for baseline scores.